Efficacy of CBT, intensified tDCS and their combination for reducing clinical symptoms and improving quality of life in social anxiety disorder with comorbid depression: a randomized controlled trial

Abstract

Background/aim

Social anxiety disorder (SAD) is a common and disabling psychiatric disorder. It is generally treated with medication and psychotherapy such as cognitive-behavioral therapy (CBT). Due to the involvement of cortical and subcortical areas in the pathophysiology of SAD, non-invasive brain stimulation techniques such as transcranial Direct Current Stimulation (tDCS) are potential adjunctive treatment options for SAD. This study aims to assess comparable efficacy of CBT, intensified tDCS, and combined CBT/tDCS on clinical symptoms and quality of life of patients with SAD and comorbid depression.

Methods

In this randomized controlled trial, 37 adults with SAD and comorbid depressive disorder were assigned into three groups: (1) CBT + active tDCS (n = 13), (2) active tDCS alone (n = 12), and (3) CBT + sham tDCS (n = 12). SAD symptoms, depressive states, quality of life and trait worry were assessed with the Liebowitz Social Anxiety Scale, Beck’s Depression Inventory, QOL questionnaire (WHOQOL-BREF), and the Penn State Worry Questionnaire respectively. The active tDCS was an intensified stimulation protocol (20 min, twice-daily sessions with 20 min intervals, 5 consecutive days) and was applied over the left dorsolateral prefrontal cortex (F3) and medial prefrontal cortex (Fpz). The CBT was provided individually based on the exposure technique at 12–20 sessions, twice a week. All clinical measures were assessed at baseline, after the intervention, and at 3-month follow-up.

Results

SAD symptoms significantly decreased after intervention and follow-up in all groups, with no significant differences between them. However, CBT + tDCS resulted in a numerically larger symptom reduction, significantly exceeding CBT + sham tDCS on the fear scale. Depressive states and trait worry significantly improved in all groups post-intervention and at the 3-month follow-up, with no between-group differences. Quality of life (total scores, physical, and psychological domains) significantly improved after the and at the 3-month follow-up only in the CBT + tDCS and tDCS-alone groups with no between-group differences.

Conclusion

Psychotherapeutic interventions​ with CBT, intensified tDCS targeting the prefrontal cortex, and the combined CBT-tDCS are effective for alleviating primary and secondary clinical symptoms in individuals with SAD. The combined CBT-tDCS intervention showed superior efficacy in reducing the primary symptoms of SAD.

Trial registration ID

IRCT20220421054607N1, registration date: 19/05/2022, available at: https://irct.behdasht.gov.ir/trial/63119.

Introduction

Social Anxiety Disorder (SAD), also known as social phobia, is the most common anxiety disorder with an early onset age and a risk factor for subsequent depression, substance abuse, and cardiovascular disease. Individuals with SAD experience fear and avoid being present in front of others and have an intense fear of social situations in which they anticipate being evaluated not only negatively [1]. The 30-day, 12-month, and lifetime prevalence of SAD are 1.3, 2.4, and 4% across all countries [2]. The prevalence of SAD in Iran ranges from 10 to 40% [3,4,5]. In general, SAD is more prevalent in females than in males [6]. SAD is often accompanied by depression [7, 8]. The lifetime comorbidity with depression is estimated to range 20–70% for patients with SAD [9]. It almost always starts first, often many years prior to the onset of depression. SAD during adolescence or young adulthood is an important predictor of subsequent depressive disorders [10]. SAD can have devastating effect on academic performance, social interactions, and job performance [11,12,13]. It can also predispose individuals to physical conditions such as insomnia, diabetes, and autoimmune diseases [14, 15], which consequently disrupt the quality of life (QOL) [16].

SAD is generally treated with medication and psychotherapy. Different types of medications can be effective in treating this disorder, including selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, beta-blockers, and anti-anxiety medications such as benzodiazepines. However, they may cause side effects, such as headaches, nausea, or difficulty sleeping. In this regard, psychotherapy methods such as cognitive-behavioral therapy (CBT) based on the exposure technique is commonly used for treatment of SAD. CBT teaches different ways of thinking, behaving, and reacting to situations to help feel less anxious and fearful. It also helps learn and practice social skills, which is very important for treating SAD. The exposure technique focuses on progressively confronting the fears underlying an anxiety disorder to help a person engage in activities s/he was avoiding. Several studies have reported the effectiveness of CBT in treatment of SAD [17,18,19,20]. Fadipe et al. in a review study furthermore reported the efficacy of CBT in improving QoL in adults with major depression [21].

Recent studies using neuroimaging techniques have highlighted the important role of cortical and subcortical areas in the pathophysiology of SAD [22,23,24], marked by hypoactivity of the dorsolateral prefrontal cortex (DLPFC), and hypersensitivity of the medial prefrontal cortex (PFC) and the amygdala [23]. In this regard, non-invasive brain stimulation techniques are a potential treatment options for SAD [24]. One of these methods is transcranial Direct Current Stimulation (tDCS) [25]. It can modulate the cortical excitability by applying a weak electrical current over the scalp through two electrode surfaces (one anode and one cathode). Anodal tDCS causes a depolarization of neurons and thus increases cortical excitability, while cathodal tDCS causes neuronal hyperpolarization and reduces cortical excitability [26, 27]. Several studies have reported the effectiveness of tDCS in treatment of major depression [28,29,30,31,32]. Jafari et al. in a sham-controlled clinical trial, reported that intensified tDCS, delivered two times a day, reduced fear and avoidance and improved QOL in SAD [33]. Sousa et al. in a report of 24 year-old female case found a reduction in the symptoms of SAD after tDCS [34].

Stein et al. [35] in a systematic review study suggested that tDCS may be more effective when used in combination with CBT and recommended large-scale clinical trials to better clarify the real effects of tDCS alone, or in combination with others. To our knowledge, there is no clinical trial on the effectiveness of tDCS combined with CBT in reducing SAD symptoms, depression, and QOL in patients with SAD in Iran. In other countries, one clinical trial was found conducted in Germany that investigated the effect of a combined CBT-tDCS intervention on major depression compared to CBT alone [36] and two studies in US combined EPR with active vs. sham tDCS to assess their effect on the alleviation of SAD symptoms [37, 38]. Moreover, the existing studies on anxiety have not simultaneously used scales that measure worry more directly, such as the Penn State Worry Questionnaire (PSWQ). This is an important issue, since some studies have argued that, because trait anxiety is multifaceted, the relationship between anxiety and another construct may not be revealed clearly unless anxiety is decomposed into some of its components such as trait worry [39]. “Trait worry is a tendency to have repetitive negative thoughts about an anticipated future event that can be difficult to remove from mind” [39]. Two studies (both in Iran) were found that examined worry (with PSWQ) along with anxiety to assess the effectiveness of tDCS; one study measured them in patients with SAD compared to sham condition [33] and other measured them in patients with generalized anxiety disorder (GAD) compared to pharmacotherapy [40]. Both reported that tDCS was effective in reducing worry. However, they did not examine the effect of tDCS combined with CBT. Considering the mentioned gaps, this clinical trial aims to investigate the effectiveness of CBT combined with tDCS in reducing SAD symptoms, depression, trait worry, and improving QOL in patients with SAD, compared to CBT and tDCS alone.

Methods

Study design and participants

This was a single-blind, randomized placebo-controlled clinical trial with 3-month follow up (prospective, ID: IRCT20220421054607N1, registration date: 19/05/2022, available at: https://irct.behdasht.gov.ir/trial/63119). The study population consists of patients with SAD referred to the Shafiyeh Clinic and the psychiatric clinic of Beheshti Hospital and those screened from university campuses in Zanjan city (Islamic Azad University of Zanjan Branch, Zanjan University of Medical Sciences, and Institute for Advanced Studies in Basic Sciences) in 2022. 78 participants were first assessed for eligibility, of whom 20 declined to participate in the study and thus were excluded. The remaining underwent structured clinical interviews based on DSM-5 (SCID-I/ SCID-II) for SAD diagnosis. Eight participants with personality disorders and five with substance use disorders were further excluded. Finally, 45 eligible patients were selected after meeting the inclusion/exclusion criteria. The participants also had comorbid depressive disorder according to the SCID-I/ SCID-II and self-report. None of them were taking any antidepressant or anti-anxiety medications during the course of experiment according to their self-report. The sample size was determined using G*power software by considering α error probability = 0.05, test power (1-β) = 80%, and effect size (f) = 0.35 [33]. Inclusion criteria were being age of 18–50 years, having at least a third-grade middle school education, amild to moderate SAD according to the SCID-I/ SCID-II and a score above 55 in the Liebowitz Social Anxiety Scale (LSAS), no severe psychiatric, cognitive, or personality disorders, no substance abuse or alcohol consumption, not receiving psychological and pharmacological brain stimulation at least one month before entering the study, and consent to participate in the study. The exclusion criteria were absence from more than two treatment sessions, suicidal thoughts during intervention according to the SCID-I/ SCID-II, pregnancy, unwillingness to continue participation, and specific contraindications for tDCS (e.g. metallic plates in the head). The participants were randomly allocated to three groups of 15, including CBT + active tDCS, tDCS alone, and CBT + sham tDCS using the block randomization method (block size of 6 with three permutations A, B, and C) in the Sealed Envelope website (https://www.sealedenvelope.com/). The allocation was done by the last author. Eight patients were excluded from the study (three from the tDCS group due to having chronic headaches or unwillingness to continue, three from the CBT group due to lack of cooperation, and two from the combined treatment group due to graduation and return to hometown). Figure 1 shows the CONSORT flowchart of the study sampling and allocation.

Measures

After obtaining written informed consent, the LSAS and Beck Depression Inventory (BDI-II) were used to measure their SAD and depression severity, respectively following the DSM-5 based structural interview with a licensed psychiatrist. A demographic form was also used to record their personal information (age, gender, educational level). Afterward, the participants completed the World Health Organization Quality of Life questionnaire (WHOQOL-BREF) and the Penn State Worry Questionnaire (PSWQ). Immediately after and three months after the last session of interventions, all groups completed these questionnaires again for post-intervention and follow-up assessments.

LSAS

The LSAS is a 24-item, self-report scale developed by Liebowitz in 1987. It measures two subscales of fear (13 items), and avoidance (11 items) in the past week. The items are first rated on a Likert Scale from 0 (none) to 3 (Severe) on the fear felt during the situations, and then the same items are rated from 0 (never) to 3 (usually) regarding avoidance of the situation. The total score is obtained by the sum of the total fear and total avoidance scores. The maximum score is 144. For SAD patients, the LSAS showed good internal consistency, test-retest reliability, convergent and discriminant validity [41]. A cut-off point of 30 shows that SAD is unlikely and a cut-off point of 60 indicates that SAD is probable. A score of 55–65 indicates mild SAD; a score of 65–80 indicates moderate SAD, a score of 80–95 shows severe SAD, and a score higher than 95 indicates very severe SAD [41]. We used the Persian version of LSAS which was validated by Khoshouei who reported its acceptable internal consistency (α = 0.95), split-half reliability, and test-retest reliability (0.83) in the Iranian population [42].

BDI-II

The BDI-II is a widely used 21-item self-report tool developed by Beck et al. in 1996 for measuring the severity of depression in adolescents and adults. The items are scored on a scale of 0–3. Total score ranges from 0 to 63. A score of 0–13 shows no depression, 14–19 shows mild level, 20–28 is moderate, and 29–63 is severe. We used the Persian version of BDI-II which was validated by Ghassemzadeh et al. [43]. They reported a high internal consistency (α = 0.87) and acceptable test-retest reliability (r = 0.74) of the Persian BDI-II.

WHOQOL-BREF

The WHOQOL-BREF is a 26-item tool that measures perceived QOL in four areas: physical health (7 items), psychological health (6 items), social relationships (3 items), and environmental health (8 items). The items are rated a five-point scale from 1 (not at all) to 5 (completely). The total score ranges from 0 to 100. The higher score indicate better perceived QOL. We used the Persian version of WHOQOL-BREF which was validated by Nedjat et al. [44]. They reported acceptable internal consistency for the domains (α ≥ 0.7 for both healthy and ill people) and acceptable test-retest reliability (r = 0.77 for physical health; r = 0.77; for psychological health, r = 0.75 for social relationships; and r = 0.84 for environmental health).

PSWQ

The PSWQ is a 16-item self-report tool designed by Meyer et al. in 1990 to measure the trait of worry in adults. The items are rated from 1 (Not at all typical) to 5 (vert typical of me). For items 1, 3, 8, 10, and 11, and 13, the scoring is reversed. The scores range from 16 to 80, with higher scores indicative of higher levels of trait worry; a score ≤ 29 indicates no trait worry; 30–52, bothered by worries but below the clinical range for worry; 52–65, some problems with worry and may benefit from treatment; and ≥ 66, Chronic worrier and in need of treatment to target this problem. Brown et al. and van Rijsoort et al. [45, 46] reported the one-factor solution of PSWQ, while Fresco et al. reported a two-factor solution [47]. We used the Persian version of PSWQ which was validated by Dehshiri et al. [48]. Their results also revealed a two-factor solution including engagement of worry (items 1–11) and absence of worry (items 12–16) and reported its high internal consistency (α = 0.88) and test-retest reliability with a one-month interval (r = 0.79).

Interventions

Participants in the tDCS-only and combined CBT + active tDCS groups received active tDCS with anodal left DLPFC (F3) and cathodal medial prefrontal cortex (Fpz) stimulation. The stimulation was delivered through a pair of saline-soaked sponge electrodes (35 cm²) for 20 min with 15 s ramp up and 15 s ramp down, twice a day with a 20-min rest interval, for 5 consecutive days making 10 number of session in total (2-mA intensity) using NeuroStim 2 device, (MedinaTeb Co., Iran). This stimulation protocol is known as “intensified” tDCS and has been used in previous studies [33, 49]. This protocol is originally taken from a physiological tDCS study in humans showing that a short (20 min) interval between stimulation sessions, induces late-phase plasticity at the physiological level [50]. The number of sessions is according to Jafari et al. [33]. Electrodes were positioned on the left DLPFC (F3) and medial PFC (Fpz), respectively, according to the EEG 10–20 International System. The tDCS was conducted by an expert in the neurotherapy department of Shahid Beheshti Hospital in Zanjan city, who was unaware of the study outcome and group allocation. To avoid confounding effects of the intervention at a circadian non-preferred time of day, which can significantly affect neuroplasticity induction, all stimulation sessions took place at fixed time of the day and participants were not under sleep pressure [51, 52].

The patients in the CBT + sham tDCS group individually received sham tDCS followed by CBT based on the exposure technique at 12–20 sessions twice a week, each for 45–90 min (Table 1). The total duration of CBT was about 2.5 months. The protocol was adapted from Leahy et al.’s study [53]. The CBT was performed by the first author with prior CBT experience under the supervision of a therapist. tDCS sessions were administered before CBT sessions. For sham tDCS, the electrical current was ramped up for 30 s to generate the same sensation as in the active condition. Then, the device turned off without the participants’ knowledge. The patients in the combined treatment group also received individual CBT at 12–20 sessions twice a week, each for 45–90 min, after receiving active tDCS. In sum, participants in the CBT+ active/sham tDCS recived five sessions of intensified tDCS (active or sham) in 5 consecutive days and then underwent 20 sessions of CBT twice a week.

Fig. 1

CONSORT Flowchart of sampling and allocation

Table 1 The protocol of CBT based on the exposure technique [48]

Data analysis

The data were analyzed using SPSS software (version 22), with the significance level set at 0.05. Descriptive statistics, including mean, standard deviation, frequency, and percentage, were employed to summarize the data. Repeated measures ANOVA was conducted to compare the means of LSAS, BDI-II, and PSWQ across groups. For the WHOQOL-BREF, multivariate ANCOVA was applied due to baseline differences. The normality of the data distribution was assessed using the Kolmogorov-Smirnov test (p > 0.05), homogeneity of variances was confirmed via Levene’s test (p > 0.05), and equality of variance-covariance matrices was checked using Box’s M test (p > 0.05). In the case of significant F values, Bonferroni-corrected post hoc t-tests were performed to evaluate both within-group and between-group comparisons.

Results

Characteristics of participants

In this study, 37 patients participated in three groups of CBT + tDCS (n = 13), tDCS (n = 12), and CBT + sham tDCS (n = 12), Table 2 shows the baseline characteristics and demographics of participants. Of 37, 26 were female (9 in the combined group, 8 in the tDCS group, and 9 in the CBT group). Most of them had a bachelor’s degree (6 in each group). The results of the chi-square test showed no significant difference among the groups in terms of educational level (p = 0.692) and gender (p = 0.9). The mean age of patients was 23.31 ± 6.49 years in the tDCS + CBT group (ranged 18–42), 23.92 ± 5.37 in the tDCS group (ranged 18–35), and 26 ± 7.78 years in the CBT group (ranged 20–42). The results of ANOVA showed no significant difference among the groups in terms of age (p = 0.576). Of 37 patients, 9 had mild depression (3 in the combined group, 4 in the tDCS group, and 2 in the CBT group), 17 had moderate depression (4 in the combined group, 6 in the tDCS group, and 7 in the CBT group), and 11 had severe depression (6 in the combined group, 2 in the tDCS group, and 3 in the CBT group). The results of the chi-square test showed no significant difference among the groups in terms of depression prevalence in any study phase (p > 0.05). Moreover, the results of ANOVA showed no significant difference in pretest scores of LSAS (p = 0.163), BDI-II (p = 0.160), and PSWQ (p = 0.361) among three groups, but the difference was significant in WHOQOL-BREF score (p = 0.002).

Table 2 Baseline characteristics of participants in three groups

Effects of interventions on social anxiety and depressive symptoms

The total LSAS score, as well as the fear and avoidance subscale scores, were analyzed with ANOVA. According to the ANOVA results presented in Table 3, the differences among the three groups over time were statistically significant only for the fear subscale score (F(2,34) = 4.15, p = 0.024). The calculated effect size (η²) indicated that 19.6% of the reduction in fear experienced during specific situations was attributable to the intervention. No significant main effects of the group were observed for the total LSAS score or the avoidance subscale score (p > 0.05). To measure the effect of the time factor, within-subject test was conducted. The ANOVA results showed a significant main effect of time and group×time interaction on LSAS total scores and both subscales (p > 0.05, Table 4). Bonferroni-corrected post hoc t-tests for between-group comparisons showed that the difference in the fear score was significantly higher in the CBT + tDCS as compared to the CBT group after the intervention (95%CI: -18.286 to -0.996; p = 0.025) and in the follow-up (95%CI: -19.839 to -2.892; p = 0.006) (Fig, 2). For within-group comparisons across time, Bonferroni-corrected post hoc t-tests showed that in all groups, there was a significant decrease in the post-intervention and follow-up scores of total LSAS score, fear, and avoidance compared to their pretest scores (p < 0.05), but the increase in their follow-up scores, compared to their posttest scores, were not significant (p > 0.05) (Fig, 2).

Fig. 2

The impact of the intervention on SAD symptoms, as measured by the LSAS (total and subscale scores), and depressive symptoms, assessed by the BDI-II, across three groups at baseline (pre-intervention), immediately after the intervention (post-intervention), and at 3-month follow-up. Note: Filled symbols/ bars indicate significant differences between pre-intervention measurements and both post-intervention and follow-up measurements in each group. Asterisks next to and above the brackets denote significant differences between the respective groups at specific time points. ns = nonsignificant. All pairwise comparisons were performed using Bonferroni-corrected t-tests. Error bars represent the standard error of the mean (s.e.m.)

For the depressive symptoms measured by the BDI-II, the ANOVA results (Table 3), showed no significant main effect of group (p = 0.429). To assess the impact of the time factor, a within-subject test was conducted (Table 4). The ANOVA results revealed a significant main effect of time (p < 0.001) but no significant interaction of group×time (p > 0.05). For the within-group comparisons, Bonferroni-corrected post hoc t-test results showed that the BDI-II scores significantly reduced after the intervention and were sustained up to 3-month follow-ups across all groups (p < 0.05). However, no significant differences were observed between post-intervention and follow-up scores in each group (p > 0.05) (Fig, 2).

Table 3 Test of between-subject effects on LSAS (SAD symptoms) and BDI-II (depressive states) scores

Table 4 Test of within-subject effects on LSAS (SAD symptoms) and BDI-II (depressive states) scores

Effects of interventions on quality of life and trait worry

Since the difference in the QOL score between groups at baseline was statistically significant, multivariate ANCOVA was used to control the effect of the pretest scores. After controlling for pretest scores (Table 5), differences in overall QOL and its four components were not statistically significant across the three groups over time (p > 0.05). The results of ANOVA (Table 6) revealed a significant main effect of time (p < 0.001) on QOL and all four components, while the interaction between time and group was significant only for psychological health and the total score (p < 0.05). In comparison to the baseline, Bonferroni-corrected post hoc tests showed significant increases in post-intervention and follow-up scores for total QOL and psychological health only in the tDCS + CBT and tDCS groups (p < 0.05). In the CBT group, these increases were significant only between pretest and posttest (p < 0.05), not at follow-up (Fig. 3). Regarding the social relationship, only the tDCS + CBT significantly improved it after the intervention, compared to the baseline (Mean difference: -2.231, 95% CI: -3.894 to -0.567, p = 0.006). No other significant changes were observed across or within groups. For environmental health, the tDCS + CBT group had significantly higher post-intervention (Mean difference: -9.231, 95% CI: -12.852 to -5.610, p < 0.001) and follow-up scores (Mean difference: -8.231, 95% CI: -13.335 to -3.127, p = 0.001) compared to pretest scores. The tDCS (Mean difference: -4.333, 95% CI: -8.102 to -0.564, p = 0.020) and CBT (Mean difference: -4.917, 95% CI: -8.686 to -1.148, p = 0.007) groups showed significant increases only in posttest scores compared to pretest scores (Fig. 3).

Fig. 3

The impact of the intervention on quality of life, as measured by the QOL across three groups at baseline (pre-intervention), immediately after the intervention (post-intervention), and at 3-month follow-up. Note: Filled symbols/ bars indicate significant differences between pre-intervention measurements and both post-intervention and follow-up measurements in each group. Asterisks next to and above the brackets denote significant differences between the respective groups at specific time points. ns = nonsignificant. All pairwise comparisons were performed using Bonferroni-corrected t-tests. Error bars represent the standard error of the mean (s.e.m.)

With respect to trait worry, the results of ANOVA (Table 7) showed no significant main effect of group (p > 0.05) and time×group interaction (p > 0.05). However, the main effect of time on trait worry total score and its components was significant for all interventions (p < 0.001) (Table 8). Bonferroni-corrected post hoc t-tests showed a significant decrease in the post-intervention and follow-up scores of trait worry and its two components compared to the pretest scores in all groups (p < 0.05) (Fig. 4). The increase in the follow-up score of total trait worry, for the engagement of worry, and absence of worry compared to the posttest score was not significant in any groups (p > 0.05).

Fig. 4

The impact of the intervention on trait worry was assessed by the PSWQ, across three groups at baseline (pre-intervention), immediately after the intervention (post-intervention), and at 3-month follow-up. Note: Filled symbols/ bars indicate significant differences between pre-intervention measurements and both post-intervention and follow-up measurements in each group. Asterisks next to and above the brackets denote significant differences between the respective groups at specific time points. ns = nonsignificant. All pairwise comparisons were performed using Bonferroni-corrected t-tests. Error bars represent the standard error of the mean (s.e.m.)

Table 5 Test of between-subject effects on quality of life

Table 6 Test of within-subject effects on quality of life

Table 7 Test of between-subject effects on trait worry

Table 8 Test of within-subject effects on trait worry

Discussion

To our knowledge, this is the first clinical trial that investigates the efficacy of CBT (with the exposure technique) combined with intensified tDCS for reducing SAD symptoms, depression, trait worry and improving QOL in patients with SAD and comorbid depressions. The tDCS group received 10 sessions of intensified tDCS with anodal and cathodal electrodes placed over the left DLPFC and medial PFC respectively (20 min, twice-daily sessions with 20 min intervals, 5 consecutive days) while the patients in the combined CBT-tDCS underwent CBT in addition to tDCS. Those in the CBT + sham tDCS only received CBT. All groups showed significant SAD symptom reduction post-intervention and at follow-up, with no significant between-group differences, although CBT + tDCS showed a numerically larger symptom reduction and significantly outperformed CBT + sham tDCS on the fear scale. Depressive states and trait worry significantly improved in all groups at both time points, again with no between-group differences. Only the CBT + tDCS and tDCS-alone groups showed significant improvements in overall QOL and in physical and psychological domains at follow-up, with no between-group differences. The efficacy of each intervention on SAD symptom reduction is discussed.

In this randomized controlled trial, our main question was to see if the combination of CBT, as a well-established intervention for anxiety disorders, and tDCS as a novel promising intervention, have superior effects on reducing clinical symptoms of patients with SAD. The results showed that all three interventions significantly reduced the SAD symptoms (fear, avoidance, total), depression, and trait worry and improved the QOL of patients. The CBT + active tDCS, resulted in numerically larger symptom reduction followed by active tDCS alone, and CBT + sham tDCS. However, the combined CBT/tDCS was particularly more effective than each intervention alone in reducing fear symptoms suggesting a potential additive effect of tDCS on CBT or vice versa. CBT, is a well-established intervention for anxiety disorders with substantial empirical support. Previous studies have demonstrated that CBT enhances the treatment of anxiety disorders through key mechanisms, including cognitive restructuring, emotion regulation, exposure therapy, and neuroplasticity. With respect to the latter, neuroimaging studies have shown that CBT can effectively strengthen amygdala connectivity with regulatory regions in patients with SAD [54, 55]. In another study on specific phobia, it decreased brain activity in cortical and subcortical structures such as prefrontal and frontal cortex and precuneus after treatment [56] and also reorganized resting-state networks, promoting plasticity that reduces fear salience and improves regulatory mechanisms in patients with panic disorder [57]. Therapeutic effects of tDCS are also attributed to functional reorganization or restoration of altered brain activity in target regions [58,59,60] that are assumed to modulate threat reactivity of cortical and subcortical regions responsible for processing threat when it comes to anxiety disorders [61, 62]. Accordingly, our finding can be explained by assuming that the combined intervention might have produced a relatively synergistic effect on symptom reduction in ASD in line with some tDCS studies [36, 63].

It is, however, noteworthy that there were no significant differences between three interventions in reducing overall SAD symptoms and depressive states and this has important clinical implications. First, tDCS over the lateral and medial PFC is an effective intervention for reducing SAD symptoms and this finding adds to previous literature on SAD [33, 64] and anxiety disorders in general [61]. Moreover, considering similar therapeutic outcomes between the tDCS group and the CBT + sham tDCS group, it can be cautiously suggested that tDCS may serve as a suitable treatment option for individuals who face difficulties with CBT or do not respond to it. Lastly, this study supports the efficacy of intensified tDCS, which has recently been used in treating SAD [33] and obsessive-compulsive disorders [49] suggesting it can be effectively used in other neuropsychiatric disorders.

In addition to SAD symptoms, the interventions also significantly reduced depressive symptoms, and trait worry, and improved QOL and all three interventions showed a relatively similar efficacy. The SAD symptoms and trait worry slightly increased after three months in all groups, while the depression level slightly increased in the CBT + tDCS group and continued to decrease slightly in two CBT and tDCS groups after three months. However, the changes in SAD symptoms and depression after three months were not statistically significant, indicating that the effects of all three interventions were sustained up to three months. Regarding the trait worry variable, the changes after three months were significant only in total score and absence of worry in the CBT + tDCS group, indicating that the effect of combined CBT-tDCS intervention on trait worry was not maintained. Regarding the QOL variable, the effects of three interventions were maintained after three months, except for environmental health in the CBT group. These findings are in line with previous studies. Previous studies have reported the effectiveness of tDCS in treatment of major depression [28,29,30,31,32], SAD symptoms [33, 34] and trait worry [33]. Huang et al. [65] demonstrated tDCS’s antidepressant efficacy (2 mA, 10 sessions, 30 min) compared to escitalopram. However, Garcia et al. [66] found no effect of a single tDCS session (2 mA, 20 min, anodal/cathodal/sham) over the left DLPFC on state anxiety, possibly due to the limited sessions or sample differences. The efficacy of CBT for SAD is well-documented [17,18,19,20, 67,68,69,70], supporting our findings. Bajbouj et al. [36] found that tDCS (2 mA, 30 min, anode F3/cathode F4) augmented CBT’s antidepressant effects in a proof-of-principle study. Cobb et al. [37] found that tDCS (1.7 mA, 20 min, active vs. sham) over the left medial PFC and right DLPFC enhanced engagement and reduced fear during exposure therapy for contamination and animal phobias, suggesting its potential for severe anxiety disorders, which aligns with our results. Conversely, McDonald et al. [38] found no benefit of tDCS combined with virtual reality exposure for public speaking SAD, potentially due to tDCS parameters or partial efficacy. Regarding the QOL, our results are consistent with the findings of other studies that applied behavioral psychotherapies [21, 69, 70] as well as tDCS studies [33, 71, 72]. In a broader picture, the improving effects of tDCS on outcome measures in our study can be explained by improved emotion regulation strategies, as shown in disorders with emotional dysregulations [73,74,75], as well as improved cognitive abilities that allow patients to gain control over emotional challenges [33, 49, 62, 76].

When interpreting the results, it’s important to consider the varying times of post-intervention assessment due to differing intervention durations. Participants in the tDCS-only intervention were evaluated after the final 5th session and again after a 3-month follow-up. In contrast, participants in the combined intervention underwent clinical assessment following the completion of CBT, which consisted of 20 sessions plus 5 daily tDCS treatments.​ This comparison is useful for understanding the therapeutic effects of intensified tDCS and also highlights the synergistic effects when tDCS is combined with CBT. Despite the novelty of the present study in combining intensified tDCS with CBT, there were some limitations and disadvantages such as low sample size (due to lack of difficulty in patient recruitment. Second, the inherently limited focality of tDCS may lead to relatively diffuse stimulation, potentially affecting additional cortical and subcortical areas. Related to this, empirical evidence supporting the subcortical effects of tDCS remains limited, and we cannot confirm that the activities of subcortical regions were influenced by our intensified tDCS. However, it is suggested that anodal left DLPFC-cathodal medial PFC tDCS may regulate subcortical regions via modulation of cortical activities [61]. Future studies should incorporate neuroimaging to better understand treatment efficacy and clarify tDCS’s unique neural contributions beyond CBT. The lack of established tDCS guidelines for anxiety disorders also necessitates further research with larger samples to optimize stimulation parameters. Despite these limitations, the study benefited from three-month outcome monitoring and inclusion of both primary and secondary clinical outcome measures.

Conclusion

The combined CBT- active intensified tDCS intervention, active intensified tDCS alone (targeting the left DLPFC and medial PFC), and individual CBT with sham tDCS have all proven effective in alleviating symptoms of SAD, depression, and trait worry in patients with SAD. Additionally, these approaches contribute to enhancing the quality of life in SAD. While no significant differences were observed among these interventions overall, the combined CBT-tDCS intervention showed superior effects on reducing fear in patients with SAD compared to the other interventions. Consequently, all three interventions are recommended for managing SAD, with the combined approach offering particular advantages in addressing fear-related symptoms.